學生英文能力及性別差異對合作閱讀策略教學成效之影響

全文

(1)國立屏東商業技術學院應用外語系(所) 碩士論文. 學生英文能力及性別差異對合作閱讀策略教學成效之影響 Effects of English Proficiency and Gender on Collaborative Strategic Reading. 指導教授：陳美貞研究生：林慧雯. 中華民國九十七年十二月.

(2) 國立屏東商業技術學院應用外語系(所) 碩士論文. 學生英文能力及性別差異對合作閱讀策略教學成效之影響 Effects of English Proficiency and Gender on Collaborative Strategic Reading. 指導教授：陳美貞研究生：林慧雯. 中華民國九十七年十二月.

(3) Effects of English Proficiency and Gender on Collaborative Strategic Reading. Advisor: Dr. Mei-Chen Chen By: Hui-Wen Lin. A Thesis Submitted to the Department of Applied Foreign Languages In Partial Fulfillment of the Requirements （. For the Degree of Master of Arts National Pingtung Institute of Commerce. Pingtung, Taiwan, R.O.C. December, 2008.

(4) 摘要合作閱讀策略教學(Collaborative Strategic Reading)建構於合作學習(Cooperative Learning)及相互教導(Reciprocal Teaching)理論之上，其教授學生四個特定的閱讀策略。然而，至今很少文獻探討此教學法，尤其在英語為外國語言的地區(EFL)，此類研究更少，因此其教學成效仍未明確。本文旨在比較此教學法與傳統文法翻譯教學法 (Grammar Translation Method)對學生閱讀說明文之成效。學生英語能力和性別差異對此教學法成效之影響亦為研究重點。本研究為實驗教學，並於實驗結束後對學生進行訪談，以增加實驗研究的信度。此研究對象為兩班國二學生(共七十八人)，並隨機對一班實施閱讀策略教學(實驗組)，而另一班則實施傳統教學(對照組)。各組分別施以六個小時，為期兩週的教學。實驗教學結束即施以閱讀評量測驗，再執行個別訪談。本研究有幾項重要發現：(一) 實驗組與對照組逹到顯著差異，前者表現優於後者；(二) 實驗組低成就學生表現明顯優於對照組同能力學生，故此類學生受惠於閱讀策略教學勝於傳統教學；(三) 相較於低成就學生，高成就及中等成就學生無此閱讀策略教學法優勢，這兩組學生無特別受惠於此類教學法；(四) 實驗組男學生表現明顯優於對照組男學生，故男同學受惠於閱讀策略教學勝於傳統教學；(五) 相較於男學生，女學生就無此閱讀策略教學法優勢，女同學無特別受惠於此類教學法；(六) 實驗組中所有的低成就受訪者皆表示他們喜歡閱讀策略教學上課方式，而只有四分之三的高成就及二分之一的中等成就受訪者表示對新教學法的認同；相對於僅一半的女學生認同此閱讀策略教學法，所有的男性受訪者皆喜歡以此方法來學習英語。最後，作者根據 i .

(5) 本研究結果提出數項教學及研究之建議，以供參考。. 關鍵詞：閱讀策略、合作閱讀策略教學法、文法翻譯教學法、英文能力、性別差異、合作學習. ii .

(6) Abstract Collaborative Strategic Reading (CSR) incorporates four reading strategies underlying the framework of cooperative learning and reciprocal teaching.. To date, little research on. CSR has been documented, particularly, in the EFL context; the effect of this instructional strategy in EFL reading is still undetermined.. The study aimed to probe the effect of CSR. on student reading comprehension of expository texts in contrast to the conventional Grammar Translation Method (GTM). approaches were further examined.. English proficiency and gender effects on the two. The study incorporated a static-group comparison. design with additional post-experimental interviews to validate the results.. Two intact. classes of 78 eighth graders were randomly assigned to the CSR instruction and the GTM instruction. The two groups received the instruction for six periods of class over a course of two weeks.. A reading comprehension test followed by individual interviews was. administered after the instruction. Important results were produced: (a) the CSR group significantly outperformed the GTM group; (b) CSR low-achievers performed significantly better than GTM low-achievers; low-achieving students benefited more from CSR instruction; (c) no CSR instructional superiority was found in high- or mid-achieving students; (d) male learners with CSR instruction significantly outperformed their male counterparts with conventional GTM instruction; males benefited more from CSR instruction; (e) no CSR instructional superiority was found in female learners; (f) all of the iii .

(7) low-achievers (100%) being interviewed reported preference over CSR in comparison with lower motivation in the high-achievers (75%) and mid-achievers (50%); particularly, 100% of the males reported positive attitudes toward this instruction in contrast to the relatively lower percentage in females (50%) on the CSR preference. Pedagogical implications and suggestions for further research were provided to conclude the study.. Key words: reading strategies, Collaborative Strategic Reading (CSR), Grammar Translation Method (GTM), English proficiency, gender, Cooperative Learning (CL). iv .

(8) Acknowledgements I would like to express the sincerest gratitude to my advisor, Dr. Mei-Chen Chen (Dr. 陳美貞), whose scholarly guidance kept me working on this thesis.. But for her. inspiration, patience, and encouragement, it would be impossible for me to complete this task.. To me, she is truly an ideal model in the academic studies and always a spiritual. guide in my life.. I am also obliged to Dr. Shwu-Mei Hwang (Dr. 黃淑眉) and Dr.. Lihung Chang (Dr. 張理宏).. With their professional knowledge, these two professors. have provided valuable suggestions to polish this study.. Finally, my heartfelt thanks are. given to my family for their supports and care all along.. It is their unconditional love that. tides me over all the difficulties in completing the thesis.. v .

(9) Table of Contents Abstract (Chinese)…………………………………………………………………………. i Abstract…………..………………………………………………………………………iii Acknowledgements………………………………………………………………………....v Table of Contents………..…………………………………………………………………vi Tables………………………………………………………………………………………ix Figures………………………………………………………………………………………x CHAPTER 1………………………………………………………………………………...1 Introduction……………………………………………………………………………1 Problem Statement……………………………………………………………….3 Purposes of the Study……...……………………………………………………..5 Research Questions………………..……..…………………………………….6 Definition of Terms……………..………………………………………………..7 CHAPTER 2………………………………………………………………………………...9 Literature Review………………………...……………………………………………9 Cooperative Learning in Reading………..……………………………………….9 Collaborative Strategic Reading……...………………………………………..12 Collaborative Strategic Reading in Relation to English Proficiency…………..20 Summary……………..……………………………………..…………………..24 vi .

(10) CHAPTER 3……………………………………………………………………………..25 Methodology…………………………………………………………………………25 Subjects…………….………………………………………………..………….26 Instruments……...……………………………………………………..………..27 Learning Materials………………………………………………………...27 Cue Cards………………………………………………………………….28 CSR Learning Log…………………………………………………………30 Reading Comprehension Test……………………………………………...32 Oral Interview Questions for the Experimental Group………………...….32 Procedures…….……………………………………………………………...…32 Data Analyses……………..………….……………………..…………………..36 CHAPTER 4……………………………………………………………………………..38 Results……………………………………………………………………..…………38 Validity and Test Reliability…………….………………………………...…….38 Inter-rater Reliability……........………………………………………..………..38 Subjects’ English Proficiency Levels…….………………………...………...…39 Results……………..………………………………………..…………………..40 Summary of Major Findings……………………………………………………48. vii .

(11) CHAPTER 5……………………………………………………………………………..49 Conclusions………………………………………………………………..…………49 Discussion on Major Findings……………..……………………………...…….49 Effects of Two Teaching Methods…………………………………………49 Effect of English Proficiency on Teaching Approach……….…………….50 Effect of Gender on Teaching Approach……………………….………….52 Perceptions of CSR………………………………………………………..53 Pedagogical Implications……........…….……………………………..………..54 Suggestions for Future Research……..…………...…………...……...……...…56 References………………………………………………………………………...……….58 Appendix A: Reading Materials…………………...………………………………………67 Appendix B: Reading Comprehension Test………...…………………………………..…76 Appendix C: Interview Questions……………………..………………………………..…85. viii .

(12) Tables Table 1 Outline of Two Most Relevant Studies on CSR with Taiwanese Subjects……….19 Table 2 Outline of the Most Relevant Studies on CSR in Relation to English Proficiency………………………………..………………………………………..23 Table 3 Design of the Study……………………………………………………………….25 Table 4 Demographic Data of Subjects…………………...……………………………….27 Table 5 Stages of a Typical Instruction for Both Groups………..………………………...34 Table 6 Descriptive Statistics for Different Levels of English Proficiency in Two Groups……………………………………………………..………………………39 Table 7 Descriptive Statistics of Reading Comprehension Test in Two Groups………...41 Table 8 Independent Samples t-test on the Test Performance of Two Groups…………….41 Table 9 Descriptive Statistics of Reading Comprehension Test in Two Groups (by English Proficiency)……………………………………….…………………..42 Table 10 Independent Samples t-test on the Test Performance of Two Groups (by English Proficiency)…………………………………………………...………43 Table 11 Descriptive Statistics of Reading Comprehension Test in Two Groups (by Gender)…………………………………………………...……………….….45 Table 12 Independent Samples t-test on the Test Performance of Two Groups (by Gender)………………………………………………………………………...45. ix .

(13) Figures Figure 1. CSR Leader Cue Card………………………………………………………….30 Figure 2. CSR Learning Log……………………………………………………………..31 Figure 3. Procedure of Implementing CSR in Group Discussion………………………..35 Figure 4. Profile Plot of Teaching Method X English Proficiency………………............43 Figure 5. Profile Plot of Teaching Method X Gender………………………………..….45. x .

(14) CHAPTER 1 Introduction The effectiveness of Collaborative Strategic Reading (CSR) in facilitating students’ reading comprehension has been highly commended in the ESL literature (e.g., Klingner & Vaughn, 2000; Klingner, Vaughn, Arguelles, Hughes, & Leftwich, 2004; Klingner, Vaughn, & Schumm, 1998; Vaughn, Klingner, & Bryant, 2001).. Basically, CSR is a teaching. approach that integrates cooperative learning (CL) into reading strategy instruction. It helps students learn four essential reading strategies to read expository texts with the assistance of the teacher and peers. These reading strategies include brainstorming and predicting (preview), monitoring understanding (click and clunk), finding the main idea (get the gist), and generating questions and reviewing key ideas (wrap up).. It is usually. implemented in small groups, with each student in the group assuming a critical role (i.e., leader, cluck expert or reporter).. Generally, CSR instruction has produced favorable. results for students with LD (learning disability) and LEP (limited English proficiency) (Kim, Vaughn, Klingner, Woodruff, Reutebuch, & Kouzekanani, 2006; Klingner et al., 1998).. Its effect on academic low-achievers cannot be overstated.. Reading is a bridge to world knowledge.. People obtain knowledge of any field,. absorb new information of any kind, and explore this changing world through the process of reading. Reading also enhances overall language proficiency of grammar, listening,. 1.

(15) speaking and writing (Anderson, 2006).. Either in a first language or foreign language. learning context, this language skill always receives great attention in the curriculum (Richards & Renandya, 2002).. It is, in particular, a skill highly valued by both teachers. and students of ESL and EFL.. Acquiring the ability to read foreign texts is important for. these learners (Richards & Renandya, 2002; Tsao, 1992). There is an urgent need for EFL teachers to teach reading skills or strategies instead of putting too much emphasis on testing (Anderson, 2003).. Strategies like CSR help. promote inter-personal interactions and learning autonomy.. Teaching students how to use. strategies should be a priority in a reading classroom (Chamot & O’Malley, 1994; Oxford, 1990).. Positive outcomes from reading strategy instruction have been generated in. previous studies (e.g., Alfassi, 2004; Deshler & Schumaker, 1993; Gajria, Jitendra, Sood, & Sacks, 2007; Klingner & Vaughn, 1996; Schunk & Rice, 1992).. Students who received. reading strategy instruction were found to outperform those who did not.. According to. Oxford (1990), use of language learning strategies can result in improved proficiency; thus, “learners need to learn how to learn, and teachers need to learn how to facilitate the process” (p. 201).. Reading is also considered as an interaction of readers’ conceptual. abilities, background knowledge and process strategies (Coady, 1979).. Hence, teachers. need to activate students’ background knowledge and provide reading strategy instruction to help learners to cope with their obstacles in reading.. 2.

(16) Unfortunately, EFL teachers in Taiwan mainly use GTM (Grammar Translation Method) to teach English reading (Chang, 2004; Liang, 1996; Tsao, 1992) and other instructional approaches are often overlooked. Vocabulary decoding, grammar analysis and bottom-up processing characterize the traditional teacher-led instruction of GTM here. As well-described by Liang (1996), a local educator and researcher, our students “learn everything about the language but not language itself” (p. 75). Reading comprehension is often neglected with too much emphasis on analyzing English grammar rules.. Students. learn to become linguists, but they hardly learn to use strategies to comprehend texts better. They might be good decoders, but there is still a long way to be good strategic readers. Moreover, fluent decoding skills do not necessarily produce better reading comprehension.. Tsao (1992) indicated that overemphasizing bottom-up processing. resulted in the fact that students only paid attention to detailed elements in the article instead of getting the whole picture of the passage. reading is to comprehend.. Note that the ultimate goal for. As indicated by Cazden (1986), the traditional teacher-led. instruction fails to achieve an actual increase in comprehension and higher level cognition. Therefore, the aim of the present study was to explore the effect of an innovative teaching strategy (i.e., CSR) on the improvement of students’ reading comprehension. Problem Statement Though CSR research has consistently yielded favorable instructional effects in the. 3.

(17) ESL literature to date (e.g., Klingner & Vaughn, 2000; Klingner et al., 2004; Klingner et al., 1998; Vaughn et al., 2001), very few studies have been conducted in the EFL context (e.g., Huang, 2004; Lee, 2003). undetermined.. The effectiveness of CSR with EFL learners still remains. Even worse, inconsistent findings were produced in the scant literature.. Huang (2004) probed the effect of an Inquiry-Based CSR instruction with senior high school EFL students and found that there was no significant difference in the reading achievement for the CSR instruction and conventional teacher-led instruction.. In contrast,. Lee (2003) detected successful CSR effect with younger EFL students (i.e., fifth graders) in studying storybooks and lyrics. Due to the limited research with the inconsistent CSR findings in the EFL context, the present study intended to compare the effects of CSR and GTM to add credence to the existing literature. In addition, the effects of English proficiency and gender on CSR instruction have not been well-explored, in particular, in the EFL classroom.. To date, the effect of English. proficiency on the effectiveness of CSR has not been confirmed yet.. Klingner et al. (2004). reported that high- and mid-achievers benefited more from CSR than from conventional teacher-centered instruction; however, Klingner et al. (1998) revealed that high-, mid- and low-achievers did not particularly benefit from either teacher-led or CSR instruction. Unfortunately, the specific effect of gender on the effectiveness of CSR has never been explored yet although previous literature has indicated gender affects strategy use (e.g., Gu,. 4.

(18) 2002; Oxford & Nyikos, 1989; Phakiti, 2003; Yu, 2006).. It calls for further research to. detect CSR effect in relation to individual differences in proficiency and gender. Purposes of the Study The present study aimed to determine the effects of proficiency and gender on the effectiveness of CSR instruction. More specifically, there were four main purposes of this. study: 1) to explore the effect of Collaborative Strategic Reading (CSR) in contrast to that of conventional Grammar Translation Method (GTM) on the reading performance of junior high school students, 2) to examine the effect of English proficiency (high-, mid- and lowability) on the instructional benefit of CSR, 3) to probe the effect of gender (male vs. female) on the instructional benefit of CSR, and 4) to find out students’ general perceptions of CSR instruction. With the purposes of study in mind, several predictions were made accordingly. Since strategy instruction had generally produced successful training effects (e.g., Alfassi, 2004; Deshler & Schumaker, 1993; Gajria et al., 2007; Klingner & Vaughn, 1996; Schunk & Rice, 1992), it was thus predicted that the CSR group would significantly outperform the conventional GTM group.. As for the ability effect, CSR was originally developed for. students with limited English proficiency (LEP) with a goal to promote their reading comprehension (Klingner & Vaughn, 2000; Klingner et al., 1998; Klingner et al., 2004; Vaughn & Klingner, 1999; Vaughn et al., 2001); therefore, it was anticipated that. 5.

(19) low-achievers would benefit more from the CSR instruction than high- or mid-achievers. When the gender effect was concerned, it was predicted that females would benefit more from the CSR instruction than males because female learners had been found better strategic users than their male counterparts (e.g., Gu, 2002; Oxford & Nyikos, 1989; Phakiti, 2003; Yu, 2006).. Lastly, positive attitudes toward CSR instruction would be. expected because learners were generally motivated by strategy training in the previous literature (e.g. C. F. Chen, 2005; M. L. Chen, 2004; Chien, 2001; Ghaith & Bouzeineddine, 2003; Huang, 2002; Liao, 2005; Vaughn et al., 2001). Research Questions 1.. Do students with Collaborative Strategic Reading (CSR) instruction perform significantly better than those with conventional Grammar Translation Method (GTM) in reading English expository texts?. 2.. Which level of learners (high-, mid- and low-achievers) benefits more from CSR instruction?. 3.. Which gender (male or female) benefits more from CSR instruction?. 4.. What are students’ general perceptions of CSR instruction?. 6.

(20) Definition of Terms Cooperative Learning (CL) Cooperative learning is defined as group learning activities in which learning is dependent on the socially structured exchange of information between learners in groups and each learner is accountable for his or her own learning and is motivated to increase the learning of others (Olsen & Kagan 1992). Collaborative Strategic Reading (CSR) Collaborative strategic reading is a set of comprehension strategies designed to improve the understanding of expository text.. It is associated with effective instruction,. such as collaborative group work, interactive dialogue and procedural strategies (Klingner et al., 2004). Grammar Translation Method (GTM) GTM instruction centers on translating the target foreign text into the mother language. Language translation and grammar analysis are vital. The primary skills to be developed are reading and writing. Little attention is given to other skills. In Taiwan, most teachers adopt this teaching method in their EFL classroom. EFL It refers to English as a foreign language, such as the English education in Taiwan, Japan and Thailand, in which English is taught as a subject in class.. 7.

(21) Expository texts Unlike a narrative text that tells a story, expository reading provides factual information.. Expository texts are written to inform, describe, explain, or persuade. They. include essays, speeches, newspaper and magazine articles, government documents, journals, directions, and lab procedures. English proficiency It is concerned about student English ability.. In the present study, it specifically. refers to the participants’ academic performance for the entire fall semester of 2007.. The. course grades of English were used to classify them as high-, mid-, and low-achievers. High-achievers referred to those who were ranked as the top-one-third learners in the class. Mid-achievers were those who were ranked as the middle-one-third learners in the class and low-achievers were bottom-one-third learners. Gender It refers to the sex classification of male and female learners. Perception It refers to students’ thoughts, ideas and attitudes here.. Perception may be influenced. by expectations, needs, unconscious ideas, values, and conflicts.. 8.

(22) CHAPTER 2 Literature Review This chapter provides general literature review related to Collaborative Strategic Reading. It consists of three sections: a) Cooperative Learning in reading, b) Collaborative Strategic Reading, and c) Collaborative Strategic Reading in relation to English proficiency.. Finally, a brief summary is given to conclude this chapter. Cooperative Learning in Reading. Collaborative Strategic Reading (CSR) is greatly associated with cooperative learning (CL).. It takes advantage of CL and integrates it into a reading strategy instruction. John. Dewey, an American educator in the 20th century and antecedent of CL, believed that students could learn better in the context of interpersonal communication and group involvement (Dewey, 1966).. The philosophy of building cooperation in learning into. regular classrooms on a systematic basis has been highly advocated ever since. In addition to cooperative learning, CSR is tied to reciprocal teaching.. CSR was first. implemented on 26 seventh and eighth graders with learning disability (LD) being instructed with CL and reciprocal teaching in the study of Klingner and Vaughn (1996), and significant improvement in reading comprehension was found in these low-achieving readers.. Reciprocal teaching, originally proposed by Palincsar and Brown (1984), aimed. to facilitate poor readers’ reading comprehension via certain comprehension monitoring. 9.

(23) and fostering activities. Four particular reading strategies involved in reciprocal teaching are summarizing, questioning, clarifying and predicting.. Students are required to practice. and model the four strategies in the first place and then take turns to be the “teacher” to lead the discussion.. Previous research has established the effectiveness of reciprocal. teaching (e.g., Palincsar & Brown, 1984; Rosenshine & Meister, 1994). There are theoretical supports derived from cognitive developmental perspective for Cooperative Learning. They are Piaget’s theory of cognition and Vygotsky’s notion of the zone of proximal development.. Piaget (1965) proposed the theory of “socio-cognitive. conflict perspective.” When people interact with others, socio-cognitive conflict occurs. It would stimulate perspective-taking ability and cognitive development which are beneficial for students’ learning.. Opportunities of interaction and discussion with. problem-solving tasks should, therefore, be provided to achieve optimal learning. Further theoretical support for CL comes from Vygotsky’s theory, “zone of proximal development” (Vygotsky, 1978).. That is, early-stage learning chiefly depends on a child’s. actual cognitive capacity and his/her potential improvement from others’ guidance. Learning is a social process that does not happen by itself but through interaction with others. In mixed-ability groups, more capable students can help improve the learning of other less capable students via social interaction. This is the true spirit of CL that students can learn better with peers’ stimulation and assistance.. 10.

(24) CL is a set of instructional techniques in which students work together in small, heterogeneous learning groups. It involves a group of three or four students internalizing certain attitudes, employing social skills, and working together to maximize their own and each other’s learning (Johnson, Johnson, & Holubec, 1994). Student-centered learning activities are thus emphasized to promote cooperation among students rather than competition. The objectives are to foster cooperation in learning, to develop critical thinking, and to develop communicative competence.. Johnson and Johnson (1999). identified five vital components of effective CL situation: positive interdependence, face-to-face promotive interaction, individual accountability, interpersonal and small-group skills, and group processing. class.. Teachers and students differ in the roles they are playing in. Students are required to learn teamwork skills and to self-evaluate in pair tasks.. Teachers act as a facilitator of learning by creating a learning environment and, meanwhile, providing broad questions to challenge student thinking. Since CL has received considerable attention, a great number of studies have explored the effects of CL instruction in relation to four language skills and student achievement (e.g., C. F. Chen, 2005; C. Y. Chen, 2004; Chen, 1999; M. L. Chen, 2005; Chien, 2004; Huang, 2007; Liang, 2003; Liao, 2005; Sachs, Candlin, Rose, & Shum, 2003; Slavin, 1983; Stevens & Slavin, 1995).. They have produced mixed findings in the comparison of. student-centered CL instruction and conventional teacher-led instruction. Particularly in. 11.

(25) the domain of reading, some studies have reported favorable CL effects (e.g., Chen, 1999; M. L. Chen, 2004; Huang, 2007; Slavin, 1983; Stevens & Slavin, 1995).. Students in the. CL group have significantly higher achievement in English reading than their counterparts in the control group with conventional instruction.. Other studies have, however,. produced contradictory results (e.g., C. F. Chen, 2005; M. L. Chen, 2005; Chien, 2004). They found that CL instruction and traditional instruction had equivalent effects in teaching English reading.. Generally, research on CL reading has generated inconsistent. findings in the ESL environment as well as in the EFL context, such as in Taiwan. Thus, the present study intended to compare CSR, which rests on the theoretical framework of CL and reciprocal teaching, with the conventional teaching method (i.e., GTM-Grammar Translation Method) to shed light on the value of cooperation in learning. Collaborative Strategic Reading Collaborative Strategic Reading (CSR) was designed to facilitate reading comprehension, in particular, for students with reading, learning, or behavior problems (Klingner et al., 2001). Basically, it integrated two main instructional approaches: reciprocal teaching and CL.. This strategic instruction was meant to teach techniques to. comprehend expository texts, to meet diverse learning needs in the classroom, and to provide students with interaction opportunities in the context of multi-ability groups (Klingner et al., 2004; Vaughn et al., 2001).. It was hoped that students with reading. 12.

(26) difficulties would progress via such instruction. CSR encompasses four reading strategies. It usually works with four students of differing abilities in small cooperative groups to assist one another in utilizing the strategies to comprehend the text.. These strategies associated with CSR instruction are. described as follows: (a). Preview (Brainstorm): This strategy aims to activate students’ background knowledge about the topic. In this pre-reading stage, students present their ideas, discuss their prior knowledge, and make predictions.. (b). Click and clunk (Locate problems): Click refers to portions of the text that make sense to the reader, and clunk means comprehension breakdowns.. In. this reading stage, students monitor their reading process and locate their own problems from the given text. (c). Get the gist (Identify the main idea): While reading, students learn to identify the most important idea from the assigned text. This strategy helps improve their understanding and memory of what they have read.. (d). Wrap-up (Integrate the text): In this post-reading stage, students review key ideas and learn to formulate questions that might be asked by the teacher. Five wh- and one how questions are involved in the final consolidating process.. 13.

(27) Becoming proficient in the use of comprehension strategies is the priority of this teaching method.. As stated by Dole (2006), “the individual texts students read are. secondary to students’ learning how to use the comprehension strategies with many different informational and expository texts” (p. 748).. Thus, with certain help from the. teacher, these strategies are continuously practiced by students in the context of cooperative learning. class as a whole.. In CSR, the four comprehension strategies are first taught to the. The teacher explains and demonstrates how each strategy can be used.. Through the Think-aloud Method, the teacher models and applies each strategy and provides students with opportunities to demonstrate as well.. When they are familiar with. the strategies, the students are divided into cooperative groups to practice them with different expository texts.. At this time, the teacher shifted the focus on to monitoring the. progress of the group studies and providing ongoing assistance as needed. CSR takes advantage of the growing knowledge base which indicates that youngsters need to be taught specific strategies to enhance their understanding of text, but should not be overwhelmed with so many strategies that they are unable to decide which ones to use (Klingner et al., 2004). Moreover, Vaughn et al. (2001) claimed that CSR provided students with the opportunity to work in small cooperative groups in which each student played a critical role associated with the effective functioning of the group and the implementation of strategies.. 14.

(28) In CSR, each student has a defined and meaningful role in the group.. To enhance. implementation of the strategies and to ensure that each student has maximum opportunities for practice, Klingner and Vaughn (1998, 1999) have suggested six possible roles in CSR implementation.. They are leader, clunk expert, gist expert, announcer,. encourager, and time keeper.. The leader is responsible for assisting the instructor to. implement CSR procedures well in each group. The clunk expert uses clunk cards to remind group members of how to deal with obstacles in the text. group members generate main ideas from each passage.. The gist expert helps. The announcer calls on different. members of the group to read and share idea to ensure everyone participates in the discussion. The encourager is responsible for encouraging everyone to participate in the group work and gives positive feedback and suggestion for improvement.. Finally, the. timekeeper watches out for the progress of time to be spent in each strategy or procedure (Klingner & Vaughn, 1998, 1999).. The six distinct roles require responsibilities and. interdependence among group members. Klingner and Vaughn pioneered a series of CSR studies with subjects of different age levels and cultural backgrounds and they consistently produced positive outcomes. Vaughn et al. (2001) issued several studies designed to enhance reading comprehension and content-area reading for students with diverse learning needs.. Development,. refinement and school model reform of CSR were made to improve the teaching strategy.. 15.

(29) First, in the development stage, Klingner and Vaughn (1996) integrated reciprocal teaching into instruction for 26 Latino middle-school students with LD and found that these students successfully applied reading strategies with some or even little adult assistance.. Then,. Klingner et al. (1998) conducted the second study with 85 fourth graders and found that CSR appeared to be feasible for use in general education elementary classrooms with heterogeneous population.. In the second stage of refinement, Klingner et al. (1998). implemented CSR on a wider scale and conducted a large project with an intensive, collaborative professional program in elementary schools for three years.. Students. fostered their reading achievements and reported favorable attitudes toward CSR.. In the. third stage of school model reform, Klingner et al. (2001) studied the integration of multicomponent reading instruction. In sum, the aforementioned studies initiated by Klingner and Vaughn generally indicated that students made significant progress in understanding both text and learning content with CSR instruction(e.g., Klingner & Vaughn, 2000; Klingner et al., 2004; Klingner et al., 1998; Vaughn et al., 2001). Furthermore, Kim et al. (2006) integrated CSR into Computer-Assisted Instruction. They examined the effects of computer-assisted comprehension practice using a self-developed computer program, Computer-Assisted Collaborative Strategic Reading (CACSR), on seventh-graders with reading difficulties. with learning disabilities participated in the study.. 16. Two teachers and 34 students. From the reading teacher’s classroom,.

(30) 12 students received the computer-assisted instruction (CACSR) and another 12 students received resource reading instruction.. From the language arts teacher’s classroom, four. students were assigned to the CACSR instruction and another six students in the comparison group received language arts instruction.. Woodcock Reading Mastery. Test-Revised Passage Comprehension (WRMT-R PC) subtest was served as the pre-test and post-test to measure students’ reading comprehension.. CACSR was implemented. through collaboration between the trained teachers and research assistants for 17-23 fifty-minute sessions. After the post-test, the CACSR students and participating teachers were interviewed.. This study revealed that students in CACSR group significantly. improved their reading comprehension.. In addition, a majority of students held positive. attitudes toward the CACSR intervention, believed that their reading had improved, and expressed a desire to learn with the program later on.. The students also indentified three. strategies of CSR (i.e., click and clunk, get the gist, and wrap-up) as helpful to their reading in the interview. Although CSR has been studied for more than one decade, very few such studies have been conducted in the area of ESL/EFL.. Particularly, little research (i.e., Huang, 2004;. Lee, 2003) has been documented to explore the effectiveness of CSR with Taiwanese EFL learners.. The two empirical studies have examined CSR as compared to the conventional. teaching method, but inconsistent findings have been generated.. 17. Huang (2004) probed.

(31) the effectiveness of an Inquiry-Based Approach via CSR technique with senior high school students.. Two intact classes, 42 students in each, took part in the empirical study for one. semester.. The experimental group received an inquiry-based reading instruction with. CSR while the control group received a traditional teacher-led instruction.. The reading. materials were selected from the textbooks. It was found that there was no significant difference between the reading achievement of the control group and that of the experimental group.. CSR and the conventional method had equivalent effects in. comprehending English reading texts.. In contrast, Lee (2003) explored the effects of. CSR and found that CSR did promote these students’ reading ability in studying storybooks and lyrics. Two intact classes of 63 fifth graders were assigned to two experimental conditions.. The students in the control group were instructed with a. teacher-directed approach, and those in the experimental group, with CSR.. The results. indicated that the CSR instruction significantly enhanced students’ reading comprehension more than the conventional teacher-led instruction.. Outline of the two most relevant. studies on CSR with Taiwanese subjects is presented in Table 1.. Due to the limited CSR. research with the inconsistent findings in the EFL context, the present study intended to compare the effects of CSR and GTM to expand the existing literature.. 18.

(32) Table 1 Outline of Two Most Relevant Studies on CSR with Taiwanese Subjects Author. Purpose of the study. Subjects. Content of teaching. Instruments. Results. background survey, interview, post-treatment questionnaire, a set of final feedback question, instructor’s observation sheet, post-reading assignment. 1. No significant differences were found in students’ academic performance with CSR instruction. 2. The CSR group reported much more progress in self-learning skills than the control group. 3. Students held favorable attitudes toward CSR.. To probe the effectiveness of an Inquiry-Based Approach via CSR techniques. 84 senior high school female students. The Far East Senior High School English Reader (textbook). Lee (2003). 1. To examine the effectiveness of CSR 2. To understand students’ reading strategies. 63 fifth graders. Storybooks, pre-test, post-test, lyrics questionnaire. 19. Huang (2004). 1. Students in the CSR context significantly outperformed their counterparts with the teacher-directed method. 2. CSR students consistently used reading strategies over time. 3. CSR elicited students’ positive reading attitudes..

(33) Collaborative Strategic Reading in Relation to English Proficiency CSR is originally designed to aid students at risk and those with LD to comprehend texts in a certain content area (Klingner & Vaughn, 1998, 1999; Klingner et al., 1998; Vaughn et al., 2001).. To date, the effect of CSR with students of differing English. proficiency cannot be determined with the existing scant literature (Klingner et al., 2004; Klingner et al., 1998). More studies were required to detect such effect of CSR in relation to individual differences. Klingner et al. (2004) reported that average- and high-achieving students in CSR instruction demonstrated greater gains of reading comprehension than low-achieving students.. Ten fifth-grade classes of a total of 221 students from five elementary schools. in the south-eastern United States participated in the study.. Five intact classes (in two. schools) were assigned to a CSR condition, and the other five, a control condition with conventional teacher-led instruction.. All of the students in both conditions engaged in. social studies and were instructed for two semesters.. Comprehension tests and prompted. Think-Aloud Strategy interviews were used to assess these students’ achievement from instruction. The results indicated that the CSR students demonstrated greater improvement in reading comprehension than the control students.. In addition, the results. of the prompted Think-Aloud Strategy interview revealed that students with LD in CSR classes showed more gains in strategic knowledge than their counterparts in the control. 20.

(34) group.. Particularly, CSR instruction benefited high- and average-learners more than. low-achieving ones, and students with LD learned more reading strategies in the CSR group than in the control group.. CSR superiority was not found in all learners, but in. specific ability groups of more proficient learners. In contrast to Klingner et al. (2004), no CSR privilege was detected in Klingner et al. (1998), which explored the effectiveness of this method in three heterogeneous, culturally and linguistically diverse, fourth-grade classes in one elementary school. The intervention condition of 85 subjects received CSR instruction; the control condition of 56 subjects received conventional instruction following the teacher’s manual.. The students. in each condition were categorized according to their exceptionality (LD or gifted students), language (English-speaking students or ESL students) and achievement (high-, averageand low-achieving students). Florida.. Both conditions learned the same material- the economy of. Eleven 45-minute instructional sessions were held constantly across both. conditions. The results indicated that the experimental students demonstrated greater gains in reading comprehension of a standard reading test and showed the same content knowledge acquisition of the learned material as the students who received traditional teacher-led instruction. It was also reported that there was no significant ability and treatment interaction or language and treatment interaction. That is, high-, average- and low-achieving students did not benefit particularly from either teacher-led or CSR. 21.

(35) instruction. Besides, both instructional methods had equivalent effects on the students who learned English as first language or second language. CSR superiority was not detected in any ability group. Note that, in Klingner et al. (2004), specific CSR instructional benefit was found in high- and average-achievers, which conflicted the result of no CSR superiority for any ability group in Klingner et al. (1998).. Such inconsistencies called for further studies on. the benefit effect of CSR on readers of differing ability.. Outline of the most relevant. studies on CSR in relation to students’ English proficiency is shown in Table 2.. 22.

(36) Table 2 Outline of the Most Relevant Studies on CSR in Relation to English Proficiency Purpose of the study. Klingner, Vaughn, Arguelles, Hughes & Leftwich (2004). To determine the relative effectiveness of CSR, in comparison with no CSR implementation, for enhancing the reading comprehension of students with LD, low-, average- and highachieving students. Klingner, Vaughn & Schumm (1998). To explore the effectiveness of a CL approach designed to foster strategic reading. 23. Author. Subjects. Content of teaching. Instruments. Results. 221 fifth graders. Social studies. Gates-MacGinite Reading Tests-Level 4, prompted Think-Aloud Strategy interview, observation, videotape, implementation validity checklists, CSR log, interview to teachers. 1. The experimental group outperformed the comparison group on all measures. 2. High- and mid-achievers benefited more from the CSR instruction than their counterparts in the control group.. 141 fourth graders. A unit pertaining to the economy of Florida. Gates-MacGinitie Reading Tests-Level 4, content measure, audiotapes. 1. Students in the experimental condition made greater gains in reading comprehension and equal gains in content knowledge. 2. There was no significant achievement level and treatment interaction, nor was language and treatment interaction..

(37) Summary In sum, regardless of successful CSR effect found in native English speakers and ESL learners, inconsistent CSR findings have been documented in the scant literature with EFL readers.. In addition, the relationship between CSR instruction and student individual. differences in ability has rarely been explored. Even worse, gender effect has never even been probed in the context of CSR strategy instruction (although previous strategy literature has generally indicated that gender affects learners’ use of strategies).. Further. empirical studies on CSR are very necessary to enlighten the issue of CSR in relation to individual differences in English proficiency and gender.. 24.

(38) CHAPTER 3 Methodology The purpose of this study was to ascertain the effect of Collaborative Strategic Reading on student reading comprehension, in contrast to conventional Grammar Translation Method.. The instructional benefit effects of CSR in relation to English. proficiency and gender were also examined.. Basically, this experimental study. incorporated a static-group comparison design. The subjects, instruments, procedures, and data analyses are described in this chapter. An overview of the entire experiment is presented in Table 3.. Table 3 Design of the Study the experimental group (n = 39). the control group (n = 39). Collaborative Strategic Reading Six periods of class over a course of two weeks Evaluated by a written test of five expository passages Interviews on individual participants. Grammar Translation Method Same Same None. 25.

(39) Subjects Two intact classes of 78 eighth graders at one junior high school in Southern Taiwan were recruited for the study. one and a half years.. They had been instructed by the same English teacher for. These two classes were randomly assigned to a CSR experimental. group and a GTM control group, and received CSR instruction and conventional GTM instruction, respectively. Each group contained 39 students (M = 19, F = 20). students had never received any CSR instruction before.. The. They were unfamiliar with this. reading strategy. In addition, English proficiency between the two groups was found to be equivalent via a statistical analysis on the students’ English term grades of the preceding semester (t = .01, p = .99 > .05).. This statistic suggested homogeneity of the two groups in their. English performance prior to the experiment.. Based on the grades, the top-one-third. students were classified as high-achievers, and the mid- and bottom-one-third, as mid- and low-achievers in their respective classes.. Note that grouping students in the experimental. CSR instruction involved an ability-grouping technique with two high-achievers, two mid-achievers and two low-achievers in a small cooperative group.. One of the two. high-achievers would be chosen to assume the role of group leader.. The participants were. not aware of such ability classifications. shown in Table 4.. The demographic data of the participants are. The heterogeneity of grouping with mixed ability levels would. 26.

(40) maximize the effect of peer tutoring (Chen & Chu, 2004; Liang, 1996; Naughton, 2006).. Table 4 Demographic Data of Subjects Group Proficiency. CSR. GTM. Gender. Male. Female. Male. Female. Total. High. 7. 6. 3. 10. 26. Mid. 8. 5. 7. 5. 25. Low. 4. 9. 9. 5. 27. Total. 19. 20. 19. 20. 78. Grand Total. 39. 39. 78. Instruments Five primary instruments were used in this study: 1) learning materials, 2) cue cards, 3) a CSR learning log, 4) a reading comprehension test, and 5) oral interview questions for the experimental group.. They were carefully reviewed by three professors specializing in. the field of English teaching and learning to ensure the content validity of these materials. Learning Materials Five expository passages were selected to be studied in both groups during the 2-week instruction (see Appendix A).. They were drawn from two reading books published by. Longman (2007) and Kang Hsuan (2008). paragraph within 130 – 150 words.. Each passage consisted of one single. The students had never learned these passages before. 27.

(41) When one passage was studied, four questions in the form of multiple choice were given in both groups to assess the students’ comprehension for the text. Basically, the questions for each passage included: (1) What does _____ (a new word or a keyword) mean?; (2) What does _____ (a pronoun) refer to in the passage?; (3) Which statement is true/false according to the passage?; (4) What is the main idea of the passage? checked immediately when the students completed the questions.. The answers were. This exercise took. place whenever a passage was learned via CSR instruction or conventional GTM instruction. Cue Cards In the CSR instruction, cue cards were used to guide the assigned role playing in a small group.. Each member assumed a constant role in the group.. The six distinct roles. in each CSR group included: 1) leader, who leads the group discussion with a focus on supervising the implementing of the four CSR strategies and involving each member in the process; 2) clunk expert, who reminds the students of the strategic steps of figuring out an unfamiliar word or a confusing part in the passage; 3) gist expert, who reminds the students of the critical steps of identifying the main idea; 4) time keeper, who prompts the timetable for the discussion progress; 5) encourager, who gives positive feedback to the group cooperation; 6) announcer, who represents the group when the teacher calls the groups back for orally reporting to the class as a whole (Vaughn & Klingner, 1999).. 28. Each role.

(42) came with a corresponding cue card that outlined the procedure to be followed in the small group.. The four strategic stages of CSR (i.e., preview, click and clunk, get the gist and. wrap up) were integrated into the group discussion procedure. Generally, the six roles were explicitly taught by the instructor and demonstrated by a model group with the cue cards in Chinese for the first two periods of class.. The cards. not only guided the students how to implement the CSR strategies in reading expository texts, but helped them stay focused on the task. Though the cards were meant to remind each small group of how to carry out the assigned roles, the students were encouraged to put them aside to proceed to a more natural discussion when they were used to the roles they were playing.. A sample of a leader cue card is presented in Figure 1.. 29.

(43) Pre-reading. During-reading. Post-reading. Preview (brainstorm and predict) S: Time keeper, please remind us of the allocating time for each stage. S: We know that today’s topic is _____. S: Who would like to share your ideas about the topic? S: Now let’s predict. Look at the title, pictures, and the headings, and think about what we might learn today. Write your ideas in your learning logs. S: Who would like to share the ideas?. Read-aloud S: Let’s take turns to read the passage.. Wrap-up (integrate the text) S: Now let’s think of some questions from what we just read. Remember to start your questions with who, when, what, where, why, & how, and write them down in your logs. S: Who would like to share the questions? S: Let’s write down what we have learned from the passage in the log. S: Announcer, could you. Click and clunk (locate problems) S: Do you understand what we have just read? If you don’t, write down your clunks in your learning logs. S: (If someone has a clunk): Clunk expert, please help us out. Get the gist (identify the main idea) S: Gist expert, please in charge of this. S: Now we will go around the group and each tells about the main idea in your own words.. summarize what we have got today? Compliments and suggestions S: Encourager, please tell us two things we did really well as a group today. S: Is there anything that would help us do even better next time?. Figure 1. CSR Leader Cue Card Adapted from Klingner and Vaughn (1999, p. 744). CSR Learning Log During the process of group discussion, the CSR students were required to record. 30.

(44) their progress of learning a certain passage in a leaning log. It was an excellent way to promote active group participation and to ensure each student really took the responsibility for his/her own learning with the instructed strategies.. A CSR learning log is given in. Figure 2.. Today’s Topic: ______________________________________ Pre-reading:. During-reading: Clunks. Preview. Date: ___________________ Post-reading: Wrap up. What do I already know about the. List the problems I have with the. List “wh-questions” for the important. topic?. passage (if any).. ideas in the passage.. __________________________. 1.. Who___________________________. __________________________. 2.. When__________________________. __________________________. 3.. What__________________________. __________________________. Where_________________________. __________________________. Why___________________________. __________________________. How___________________________. What do I want to learn? or. What have I learned from the passage?. What do I predict I will learn? __________________________. _______________________________. __________________________. _______________________________. __________________________. _______________________________. __________________________. _______________________________. __________________________. _______________________________. __________________________. _______________________________. Figure 2. CSR Learning Log Adapted from Klingner & Vaughn (1999, p. 291). 31.

(45) Reading Comprehension Test Five expository passages with a total of 20 multiple-choice questions were selected to assess the reading progress in both instructional conditions (see Appendix B). They were of equivalent difficulty level as those used during the instruction.. The same types of. questions as those used during the exercises were also given in the reading comprehension test.. Generally, four questions were set up for each passage with each question. accounting for five points in the test. Oral Interview Questions for the Experimental Group Oral interviews concerning attitudes toward CSR instruction were held on 12 selected subjects consisting of four high-, four mid-, and four low-achievers. Four interview questions were given and all of them were semi-structured to elicit more flexible responses (see Appendix C).. These interview questions were as follows:. 1. Do you like to learn English with the method of CSR?. Why or why not?. 2. Do you think CSR facilitates the comprehension of the passages? 3. Did you encounter any difficulties during the group discussion with the CSR strategy? 4. Which teaching method do you prefer, CSR or conventional GTM?. Why?. Procedures When the permission of conducting an experimental study was obtained from the school authorities of the participants, two intact classes were assigned randomly to an. 32.

(46) experimental group and a control group.. Five expository texts were taught to both groups. in six 45-minute classes over a course of two weeks.. The experimental group received. the CSR instruction, the control group, the conventional Grammar Translation Method. The control group was taught prior to the experimental group by one English teacher on the same days during the entire instructional period.. The typical teaching progress of. both groups for each class is presented in Table 5. Basically, each 45-minute instruction was broken into three stages (i.e., preparatory, instructing, and evaluating) to study an expository passage.. A detailed description of the implementing of the four CSR reading. strategies during group discussion is given in Figure 3.. 33.

(47) Table 5 Stages of a Typical Instruction for Both Groups Stage Preparatory Stage. Instructing Stage. Evaluating Stage. Experimental group (CSR). Control group (GTM). Time (mins). Review the four strategies of CSR. Provide prior knowledge of the new reading passage.. 5. 1. Group discussion: (1) Preview (2) Click and clunk (3) Get the gist (4) Wrap up 2. Review activities: Orally present group products. 1. Teach vocabulary 2. Analyze the sentence structures and study the involved grammar 3. Translate the text into Chinese 4. Read the whole passage aloud 5. Review important grammar 6. Summarize the passage. 30. Four multiple-choice questions for the studied passage.. Same. 10. 34.

(48) Pre-reading Preview 1. Brainstorm: What do we already know about the topic? 2. Predict: What do we think we will learn about the topic when we read the passage?. During-reading Click & clunk (locate problems) 1. Are there any parts or words that are hard to understand (clunks)? 2. How can we fix the clunks? Fix-up strategies: (a) Reread the sentence and look for key ideas to help you understand the word. (b) Reread the preceding and following sentences to look for clues. (c) Look for a prefix or suffix in the word. (d) Break the word apart and look for smaller words.. Get the gist (identify the main idea) 1. What is the most important person, place, or thing? 2. What is the most important idea about the person, place, or thing?. Post-reading Wrap-up (integrate the text) 1. Ask 5-W and 1-H questions: What questions would assess whether we understand the most important information in the passage? Can we answer the questions? 2. Review: What did we learn? Figure 3. Procedure of Implementing CSR in Group Discussion Adapted from Klingner and Vaughn (1999, p. 740). 35.

(49) Immediately after the 2-week instruction, a test consisting of five passages was administered to both groups to assess the students’ reading achievements via the two different methods.. These passages were different from those used in the training, but of. the same difficulty level.. Finally, to further examine student responses to the CSR. instruction, an open-ended oral interview was then conducted.. Twelve students of. differing ability levels received such interviews individually. These interviews took 70 minutes to complete. Data Analyses The collected data included: 1) a reading comprehension test of five expository texts, and 2) post-experimental interviews on selected students concerning attitudes toward CSR intervention.. The data analyses, thus, were both quantitative and qualitative.. Two experienced English teachers were invited to grade the assessment test.. The. reliability coefficient (r = 1.0, p = .000 < .01) found in the study indicated a full scoring consistency between the two judges. The scores were later keyed in a computer file for further statistical analyses via the SPSS program.. To probe the effect of CSR in contrast. to a conventional GTM on comprehending expository texts, an Independent Samples t-test was used to compare the performance of the two conditions on the reading comprehension test.. To probe the instructional benefit effect involving English proficiency, three. Independent Samples t-tests were conducted on the three-ability levels (high, mid, low) for. 36.

(50) their respective performances in the two instructions.. Similarly, another two Independent. Samples t-tests were executed to determine the instructional benefit effect involving gender (males and females).. The level of statistical significance for all of the tests was set at .05.. In addition to these statistical tests, the students’ responses regarding CSR instruction in the interviews were recorded and transcribed verbatim for further analyses.. 37.

(51) CHAPTER 4 Results This chapter reports the results in regard to the research questions addressed in chapter one. Validity and reliability of the study, inter-rater reliability, and subjects’ English proficiency levels are presented before detailed descriptions of the results.. The. qualitative findings of students’ attitudes toward CSR are also illustrated. Validity and Test Reliability As mentioned in the previous chapter, before the experiment, all the learning materials, the reading comprehension test and the interview questions were carefully reviewed by the subjects’ English teacher and three professors specializing in English teaching. According to their suggestions, revisions were made to ensure the content validity of and consistency between the learning materials and the evaluation instrument. Kuder-Richardson 20 Procedure was applied to examine the internal consistency of the test. The test reliability (r = .85) was found to be satisfactory. Inter-rater Reliability Two independent scorers were invited to grade the reading comprehension test on the subjects.. Pearson correlation coefficient procedure was performed on the scores obtained. from the graders to generate an inter-rater reliability for the test.. The reliability. coefficient (r =1, p = .000 < .01) indicated full scoring agreement between the two judges.. 38.

(52) Subjects’ English Proficiency Levels Based on their academic reading achievement (i.e., English term grades of the preceding semester), the subjects in both groups were divided into high-, mid-, and low-achievers.. Specifically, the top-one-third learners in each class were classified as the. high-achievers, and their middle- and bottom-one-third counterparts, as mid- and low-achievers.. The splitting points of the three levels on both groups were 84 and 57.. The subjects were unaware of such classifications.. The descriptive statistics of the three. levels of English proficiency for the CSR group and the GTM group are shown in Table 6.. Table 6 Descriptive Statistics for Different Levels of English Proficiency in Two Groups Score Range. English Proficiency. Numbers. Average. (N). Scores. 84-96 (n=13). 26. 90.51. 59-83 (n=13). 63-81 (n=12). 25. 71.42. 19-52 (n=13). 0-56 (n=14). 27. 33.37. CSR group. GTM group. High. 84-97 (n=13). Mid Low. 39.

(53) Results The statistical results are presented according to the sequence of the research questions. In the first place, the effect of the teaching approaches (CSR vs. GTM) on student reading performance is reported.. The results of the two teaching methods in. relation to proficiency and gender are then addressed.. At last, the interview findings of. the students’ attitudes toward CSR are revealed. Interpretations of all results are also provided. Research question 1: Do students with Collaborative Strategic Reading (CSR) instruction perform significantly better than those with conventional Grammar Translation Method (GTM) in reading English expository texts? Descriptive statistics and a statistical analysis for the performance on the reading comprehension test in both groups are displayed in Table 7 and 8. The mean of the CSR group (M = 58.08) was greater than that of the GTM group (M = 46.41).. An Independent. Samples t-test was then executed, and a significant mean difference in the two groups was found (t (76) = 2.14, p = .036 < .05).. The subjects with CSR instruction significantly. outperformed those with conventional GTM method on the test.. Namely, CSR method. significantly promoted junior high school students’ comprehension of English expository texts.. 40.

(54) Table 7 Descriptive Statistics of Reading Comprehension Test in Two Groups Research Groups. N. M. SD. CSR group. 39. 58.08. 24.00. GTM group. 39. 46.41. 24.22. Table 8 Independent Samples t-test on the Test Performance of Two Groups t. df. Sig.. 2.14. 76. .036*. *p < .05.. Research question 2: Which level of learners (high-, mid- and low-achievers) benefits more from CSR instruction? As shown in Table 9, the means of the reading comprehension test in the three CSR ability groups were all greater than those in the GTM contrastive groups.. An Independent. Samples t-test was then conducted to examine the mean difference in the two groups for each proficiency level. It was found that only the mean difference of the low-achieving level reached statistical significance (t (25) = 3.04, p = .006 < .05), as indicated in Table 10. The low-achievers in the CSR group performed significantly better than those in the GTM. 41.

(55) group.. This suggested that low-achieving students benefited more from CSR instruction. than from GTM instruction.. However, such CSR instructional advantage was not found. in their high- or mid-achieving counterparts.. A significant mean difference in the two. instructional groups was not detected in either of the two proficiency levels (high-achievers: t (24) = .20, p = .841 > .05; mid-achievers: t (23) = 1.54, p = .137 > .05). This implied that the two teaching approaches had equivalent effects in teaching high- or mid-achievers; no particular instructional superiority was thus found for these learners.. A profile plot of. teaching method by English proficiency is presented in Figure 4. To sum up, low-ability learners with CSR instruction significantly outperformed those with GTM instruction, while no significant instructional difference was found in high- or mid-ability learners.. It could be concluded that low-achievers benefited more from the. CSR instruction than high- or mid-achievers. Table 9 Descriptive Statistics of Reading Comprehension Test in Two Groups (by English Proficiency) Research Groups. N. M. SD. CSR. High = 13 Mid = 13 Low = 13. 71.15 58.08 45.00. 25.83 20.48 19.15. GTM. High = 13 Mid = 12 Low = 14. 69.17 47.31 26.07. 22.85 14.81 12.12. 42.

(56) Table 10 Independent Samples t-test on the Test Performance of Two Groups (by English Proficiency) Levels. t. df. Sig.. High. .20. 24. .841. Mid. 1.54. 23. .137. Low. 3.04. 25. .006*. *p < .05.. 80 71.15. Performance of Test. 70 60. 69.17 58.08. 50 45.00. 47.31. 40. CSR GTM. 30 26.07 20 10 0 High. Mid. Low. English Proficiency. Figure 4. Profile Plot of Teaching Method X English Proficiency. 43.

(57) Research question 3: Which gender (male or female) benefits more from CSR instruction? Table 11 lists the mean scores and standard deviations of the male and female students’ performance on the reading comprehension test in both of the CSR and GTM groups.. To detect instructional advantage involving gender, an Independent Samples. t-test was executed to examine if there was a significant mean difference in the male learners in both groups, and another t-test, in the female learners in both groups.. As. shown in Table 12, a significant difference was detected between the two male groups on the reading comprehension test (t (36) = 2.39, p = .022 < .05).. That is, the male learners. under CSR (M = 59.47) significantly outperformed their male counterparts under GTM (M = 40.79), which suggested the male learners benefited more from the innovative CSR than from the conventional GTM.. However, no such significant mean difference was found in. the female learners of the two groups (t (38) = .66, p = .515 > .05), which suggested females did not particularly benefit from either of the two approaches. In sum, male learners significantly benefited more from CSR while such instructional superiority was not found in female learners in the present study. the CSR instruction than females.. Males generally benefited more from. A profile plot of teaching method by gender is. presented in Figure 5.. 44.